sign in sign up
<<
Home , Data science

Science Initiative 1

3 credits in data and , 1 credit in societal implications and opportunities, Data Science Initiative 1 elective credit to be drawn from a wide range of focused applications or deeper theoretical exploration, and 1 credit capstone experience. Director We also offer an option as a 5-th Year Master's Program if you are an Sohini Ramachandran undergraduate at Brown. This allows you to substitute maximally 2 credits Direfctor of Graduate Studies with courses you have already taken. Samuel S. Watson Master of Science in Data Science Brown University's Data Science Initiative serves as a campus hub Semester I for research and education in data science. Engaging partners across DATA 1010 Probability, , and Machine 2 campus and beyond, the DSI 's mission is to facilitate and conduct both Learning domain-driven and fundamental research in data science, increase data DATA 1030 Hands-on Data Science 1 fluency and educate the next generation of data scientists, and ultimately DATA 1050 Data Engineering 1 explore the impact of the data revolution on culture, society, and social justice. We envision our role in the university and beyond as something to Semester II build over time, with the flexibility to meet the changing needs of Brown’s DATA 2020 Statistical Learning 1 students and research community. DATA 2040 and Special Topics in Data 1 The Master’s Program in Data Science (Master of Science, Science ScM) prepares students from a wide range of disciplinary backgrounds DATA 2080 Data and Society 1 for distinctive careers in data science. Rooted in a research An appropriate 1000-level or 2000-level course to be determined 1 collaboration between four strong academic departments (Applied by the student and approved by the program advisor. Possible , Biostatistics, , and Mathematics), the courses could range from advanced mathematical methods to Master's Program offers a unique and rigorous education for people very specific applications of data science. building careers in data science and/or management. Summer For additional , please visit the initiative's website: http:// DATA 2050 Data Science Practicum 1 1 dsi.brown.edu/ Total Credits 9 1 Data Science Graduate Program For their capstone experience, students will work on a project with real data, potentially in any one of the areas covered by the elective Master of Science in Data Science course. A faculty member from one of the four departments will The Data Science Initiative at Brown offers a new master's program (ScM) oversee the capstone course, although each student may collaborate that will prepare students from a wide range of disciplinary backgrounds with an additional faculty member, postdoc, or industry partner on his/ for distinctive careers in Data Science. Rooted in a research collaboration her project. among four very strong academic departments (Applied Mathematics, For more information on admission and program requirements, please visit Biostatistics, Computer Science, and Mathematics), the master's program the following website: will offer a rigorous, distinctive, and attractive education for people building careers in Data Science and/or in Big . The program's https://www.brown.edu/academics/gradschool/programs/data-science main goal is to provide a fundamental understanding of the methods (https://www.brown.edu/academics/gradschool/programs/data-science/) and algorithms of Data Science. Such an understanding will be achieved Courses through a study of relevant topics in mathematics, statistics and computer science, including , , security and privacy, DATA 0080. Data, Ethics and Society. visualization, and data management. The program will also provide A course on the social, political, and philosophical issues raised by experience in important, frontline data-science problems in a variety the theory and practice of data science. Explores how data science is of fields, and introduce students to ethical and societal considerations transforming not only our sense of science and scientific knowledge, surrounding data science and its applications. but our sense of ourselves and our communities and our commitments concerning human affairs and institutions generally. Students will examine The program's course structure, including the capstone experience, will the field of data science in light of perspectives provided by the philosophy ensure that the students meet the goals of acquiring and integrating of science and technology, the sociology of knowledge, and science foundational knowledge for data science, applying this understanding in studies, and explore the consequences of data science for life in the first relation to specific problems, and appreciating the broader ramifications half of the 21st century. Fulfills requirement for Certificate in Data Fluency of data-driven approaches to human activity. Moreover, our strong Fall DATA0080 S01 17954 TTh 10:30-11:50(13) (D. Hurley) industry partnerships will help you better learn about industry's needs and directions, and will expose you to novel and unique opportunities. DATA 0200. Data Science Fluency. In addition, several professors from all across the different department's As data science becomes more visible, are you curious about its unique groups work closely with industry (regional and beyond) and the amalgamation of computer programming, statistics, and visualizing government, so you will be able to sharpen your skills here on problems or storytelling? Are you wondering how these areas fit together and that bring research ideas and methods to bear on problems of practical what a data scientist does? This course offers all students regardless value. of background the opportunity for hands-on data science experience, The program will be conducted over one academic year plus one summer, following a data science process from an initial research question, through with the option for an additional pre-program summer for students who , to the storytelling of the data. Along the way, you will learn lack one or more of the basic prerequisites. The regular program includes about the ethical considerations of working with data, and become more two semesters of coursework and a one-summer (5- 10 week) capstone aware of societal impacts of data science. Course does not count toward project focused on data analysis in a particular application area. CS concentration requirements. Spr DATA0200 S01 26425 TTh 2:30-3:50(11) (L. Clark) There are nine credits unites required to pass the program: four in each of the academic year semesters, and one (the capstone experience) in the summer. The nine credit-units divide as follows: 3 credits in mathematical and statistical foundations,

Data Science Initiative 1 2 Data Science Initiative

DATA 1010. Probability, Statistics, and Machine Learning. DATA 2020. Statistical Learning. An introduction to the mathematical methods of data science through A modern introduction to inferential methods for and a combination of computational exploration, visualization, and theory. statistical learning, with an emphasis on application in practical settings Students will learn scientific basics, topics in numerical linear in the context of learning relationships from observed data. Topics will algebra, mathematical probability (probability spaces, expectation, include basics of , variable selection and dimension conditioning, common distributions, law of large numbers and the central reduction, and approaches to nonlinear regression. Extensions to other limit theorem), statistics (point estimation, confidence intervals, hypothesis data structures such as longitudinal data and the fundamentals of causal testing, maximum likelihood estimation, density estimation, bootstrapping, inference will also be introduced. and cross-validation), and machine learning (regression, classification, and Spr DATA2020 S01 25713 TTh 10:30-11:50(09) (. DeVito) , including neural networks, principal component analysis, and ). DATA 2040. Deep Learning and Special Topics in Data Science. A hands-on introduction to neural networks, , Fall DATA1010 S01 18019 MWF 11:00-11:50(14) (S. Watson) and related topics. Students will learn the theory of neural networks, Fall DATA1010 S01 18019 MWF 10:00-10:50(14) (S. Watson) including common optimization methods, activation and loss functions, DATA 1030. Hands-on Data Science. regularization methods, and architectures. Topics include model Develops all aspects of the machine learning pipeline: data acquisition and interpretability, connections to other machine learning models, and cleaning, handling missing data, exploratory data analysis, visualization, computational considerations. Students will analyze a variety of real-world , modeling, interpretation, presentation in the context problems and data types, including image and natural language data. of real-world datasets. Fundamental considerations for data analysis Spr DATA2040 S01 26298 TTh 1:00-2:20(08) ’To Be Arranged' are emphasized (the bias-variance tradeoff, training, validation, testing). DATA 2050. Data Science Practicum. Classical models and techniques for classification and regression are The capstone experience is a hands-on thesis project that entails an in- included (linear and with regularization, support vector depth study of a current problem in data science. Students will synthesize machines, decision trees, random forests, XGBoost). Uses the Python their knowledge of probability and statistics, machine learning, and data data science ecosystem (e.g., sklearn, pandas, matplotlib). Prerequisites: and computational science. A faculty member from one of the four core A course equivalent to CSCI 0050, CSCI 0150 or CSCI 0170 are strongly DSI departments (Applied Mathematics, Biostatistics, Computer Science, recommended. Mathematics) will oversee the capstone course. Students may collaborate Fall DATA1030 S01 17952 TTh 10:30-11:50(13) ’To Be Arranged' with an additional faculty member, postdoc, or industry partner on projects. DATA 1050. Data Engineering. DATA 1010 and DATA 1030 are recommended pre-requisites. Provides an introduction to computer science and programming for data Fall DATA2050 S01 17955 Arranged ’To Be Arranged' science. Coverage includes data structures, algorithms, analysis of DATA 2080. Data and Society. algorithms, algorithmic complexity, programming using test-driven design, A course on the social, political, and philosophical issues raised by code organization, and version control. Additional topics include SQL, no- the theory and practice of data science. Explores how data science is SQL, and graph , , and web technologies. transforming not only our sense of science and scientific knowledge, Fall DATA1050 S01 17953 TTh 1:00-2:20(08) (S. Watson) but our sense of ourselves and our communities and our commitments DATA 1150. Data Science Fellows. concerning human affairs and institutions generally. Students will examine Data science is growing fast, with tools, approaches, and results evolving the field of data science in light of perspectives provided by the philosophy rapidly. This course is for students with some familiarity with data science of science and technology, the sociology of knowledge, and science tools and skills, seeking to apply these skills and teach others how to studies, and explore the consequences of data science for life in the first implement and interpret data science. Working in conjunction with a half of the 21st century. faculty sponsor, this course teaches students communication skills, how Spr DATA2080 S01 26278 Th 4:00-6:30(17) (D. Hurley) to determine the needs (requirements) for a project, and how to teach DATA 2110. Topics in Econometrics. data science to peers. These valuable agile skills will be an incredible This course will begin with a survey of the literature on identification advantage moving forward in your professional development. Interested using instrumental variables, including identification bounds, conditional students should complete the initial application form below to express moment restrictions, and control function approaches. The next part of interest. Override requests will be granted only to students by instructor class will cover some of the theoretical foundations of machine learning, approval. including regularization and data-driven choice of tuning parameters. We https://docs.google.com/forms/d/ will discuss in some detail the canonical normal means model, Gaussian e/1FAIpQLSciIksIFrviEDy4mGpwRa3ekCpozL7-9UtR51bkgKL2gWGqpw/ process priors, (empirical) Bayes estimation, and reproducing kernel viewform?usp=sf_link Hilbert space norms. We will finally cover some selected additional topics Fall DATA1150 S01 18014 TTh 2:30-3:50(12) (L. Clark) in machine learning, including (deep) neural nets, text as data (topics models), multi-armed bandits, and . DATA 1200. Reality Remix - Experimental VR. This course pursues collaborative experimentation with virtual and augmented reality (AR and VR). The class will work as a team to pursue research (survey of VR/AR experiences, scientific and critical literature review), reconnaissance (identifying VR/AR resources on campus, in Providence and the region), design (VR/AR prototyping). Research findings are documented in a class wiki. The course makes use of Brown Arts Initiative facilities in the Granoff Center where an existing VR laboratory will be expanded through the course of the semester based on student needs. Class culminates in the release the class wiki as a resource for the Brown community.

2 Data Science Initiative